Role: Data Scientist – Modeller
Primary Location: Manchester, UK
Employee Status: Permanent (Full Time)
Salary: Competitive
The Company
Hello Soda is an international big data and text analytics business, headquartered in the heart of Manchester, with offices in Bangkok and Austin. We develop and sell software to businesses which empowers consumers to leverage their digital footprint to gain access to more personalised services, in turn helping businesses to verify identity and increase customer acquisition, reduce fraud, and personalise the user experience.
We are a small but fast-growing integrated team of management, sales, marketing, design, data science, data analytics, and development. Innovation is at the core of what we do, and we are revolutionising the way that social and unstructured data is being utilised across a range of industry verticals.
The Role
Hello Soda are looking for a Data Scientist with specific experience in modeling, feature generation and feature extraction who is used to implementing complex analytics for meaningful insights, of prototyping new models, and operating over a variety of business challenges in multiple sectors.
You will be required to perform exploratory research and help develop data processes in data analysis, predictive data modelling, time series analysis and anomaly detection or visualisation for a wide range of disparate datasets.
You will join a highly skilled team of Data Scientists, Analysts and Software Engineers, with a passion for developing solutions in modern technologies, to contribute analysis and models to obtain insight from a variety of unstructured and semi-structured data.
Responsibilities
- Carry out exploratory data science projects and building predictive models and scorecards for a range of verticals
- Identify new predictive insights and features from a variety of data sources
- Apply a wide range of modelling techniques to tackle unsupervised, semi-supervised and supervised learning problems
- Data and variable re-structuring, pre-processing and transformation across a wide range of data sources and API endpoints
- Adapting modelling techniques for both small and large datasets (bootstrap methods, cross-fold validation, etc)
- Validation and monitoring of model performance
Skills
- Ability to model solutions to problems in innovative and original ways
- Experience with all stages of a data science project, from importing and cleaning data from a range of sources and formats, feature engineering and model validation using R
- Self motivation to drive the creation of data models forward and make them accurate and meaningful to others
- Excellent communication skills, including data visualisation and reporting.
- Strong technical skills, proficiency in programming in R is a must
- Meticulous approach to data-handling and cleansing and an attention to detail
The Person
- Passion for research and development
- Demonstrates a high level of initiative and has an enquiring mind
- Wants to be part of a high growth start-up company with global ambitions
Desirable:
- Experience with NoSQL (for example, MongoDB) databases
- Experience with ETL processes and AWS
- Previous experience in a B2B environment
- Knowledge of building production ready modelling algorithms
- Experience with technologies such as Docker, Spark and/or Shiny
- Significant practical application of multi-variate techniques, such as; Decision Trees, Random Forest, Naïve Bayes, Clustering, Gaussian Processes, Generalised Regression, etc across a range of business requirements